Amazon: Nova 2 Lite

Amazon: Nova 2 Lite is a text generation model from Amazon, part of the Nova family. It is described by the provider as a fast, cost-effective reasoning model aimed at everyday workloads. The model was released by the provider on 2 December 2025. It sits in the text category on api14, and its model id is amazon/nova-2-lite-v1.

Nova 2 Lite accepts text, images, video and files as input, and it produces text as output. That combination makes it a multimodal model on the input side while remaining text-only in what it returns. The context window is 1,000,000 tokens, which is large enough to hold very long documents, extended conversations or sizeable media-derived inputs in a single request. The maximum output is 65,535 tokens, so responses can be long as well. Supported request parameters include include_reasoning, max_tokens, reasoning, stop, temperature, tool_choice, tools, top_k and top_p.

Because it can read images, video and files alongside text, practical uses follow from what you can feed it. You can send a document and ask for a summary or for answers drawn from its contents. You can pass images and ask for descriptions or for text extracted from what is shown. You can supply video and ask for a textual account of it. The reasoning parameters and the large context window suit multi-step tasks where the model needs to work through a problem before answering, and the tool parameters let it call functions you define, which is useful for workflows that need to fetch data or trigger actions. Everyday workloads such as drafting, classification, question answering over long inputs and assistant-style chat are all reasonable fits.

You call it through api14's OpenAI-compatible API using the model id amazon/nova-2-lite-v1. Because the interface follows the OpenAI format, you can point an existing OpenAI client at api14 and set that id as the model. The request body accepts the parameters listed above: max_tokens to cap the response length, temperature, top_p and top_k to shape sampling, stop to define stopping sequences, and tools together with tool_choice to enable function calling. The reasoning and include_reasoning parameters control whether reasoning is requested and whether it is included in the response. The page shows a live price table for current rates.

One thing to keep in mind is that output is text only, so if you need generated images, audio or other output types, this model will not provide them. The 65,535 token output cap also means very long generations have to be split across requests. Within those bounds, Nova 2 Lite is a reasonable default for text tasks that involve large inputs or mixed media inputs.